A GPU-Outperforming FPGA Accelerator Architecture for Binary Convolutional Neural Networks.
Yixing LiZichuan LiuKai XuHao YuFengbo RenPublished in: ACM J. Emerg. Technol. Comput. Syst. (2018)
Keyphrases
- convolutional neural networks
- field programmable gate array
- real time
- hardware implementation
- parallel implementation
- hardware architecture
- parallel architecture
- hardware design
- fpga implementation
- parallel computing
- software implementation
- pipelined architecture
- hardware architectures
- dedicated hardware
- xilinx virtex
- low cost
- fpga technology
- heterogeneous computing
- compute intensive
- systolic array
- fpga device
- high speed
- management system
- reconfigurable hardware
- embedded systems
- efficient implementation
- software architecture
- signal processing
- multi class
- gpu implementation
- multi core processors
- parallel architectures
- single instruction multiple data
- general purpose
- convolutional network